The Repeat Expansion Diseases are genetic disorders that result from an expansion of a specific tandem repeat array. The consequences of expansion depend on the location of the array in the affected gene, and in many cases on the sequence of the repeat. Where expansion occurs in an open reading frame, the connection between expansion and disease symptoms is relatively straightforward. However, where the repeat is located outside of the open reading frame, the connection between expansion and disease symptoms is less clear. Disorders in this category include Fragile X syndrome (FXS) and Friedreich's ataxia (FRDA). FXS is the most common heritable cause of mental retardation. Carriers of FXS "premutation alleles" show a much higher incidence of ovarian and cerebellar dysfunction than individuals with the "full mutation". FXS is caused by expansion of a CGG/CCG-repeat in the 5' UTR of the FMR1 gene. This expansion has 2 apparently paradoxical effects on transcription. At "premutation" lengths, it leads to an increase in FMR1 transcription, while in the full mutation it leads to hypermethylation of the promoter, and a dramatic decrease in transcription. In addition to the effects on transcription, expansion leads to stalling of 40S ribosomal subunits and a subsequent decrease in the efficiency of translation. FRDA is a degenerative disease associated with cerebellar dysfunction, cardiomyopathy, and diabetes. It is caused by GAA/TTC-repeat expansion in the first intron of the frataxin gene. We are interested in both the mechanism of expansion and the consequences of expansion in these disorders. Mouse models of repeat expansion: We have previously shown that the FXS repeats are not intrinsically prone to expansion in transgenic mice. This suggests that sequence context or differences in trans-acting factors might play a role in expansion in humans. We have also shown that the failure to detect expansion in our CGG/CCG-containing mice is not due to the absence of maternal imprinting, since CGG/CCG-repeats embedded in a transgene that confers a cis-acting imprinting signal also did not expand. In addition, failure to see expansion in mice is not due to hypervigilant p53 DNA damage surveillance or repair since mice nullizygous for p53 do not show expansions either. We have now shown that mice nullizygous for the Werner's and Bloom's syndrome helicases also show no evidence of expansion. This is somewhat surprising given that these proteins have been proposed to act to unwind secondary structures such as the FXS tetraplexes during replication or repair. Furthermore, XPG, a flap endonuclease homolog of rad27p, a protein that reduces expansion mutations in yeast, also shows no expansions. These data suggest that perhaps sequence context is a more important determinant of repeat instability. Since mice and humans are syntenic in the region of the X chromosome containing the FMR1 gene, it is hoped that a "knock-in" mouse containing a long CGG/CCG-tract in the 5' UTR of fmr1 should have the appropriate cis-acting sequences for expansion. The generation of mice from ES cells containing a targeted insertion of ~120 CGG/CCG-repeats is currently underway in collaboration with Dr Robert Nussbaum at NHGRI. FMR1 gene regulation: We have extended our previous work on the regulation of the FMR1 gene to show that the promoter is intrinsically bent, with the bend center located just 5' of the transcriptional start site. This bending is exacerbated by extrinsic bending caused by USF1/USF2 and NRF-1 which we have shown to be the transcription factors most important for FMR1 regulation. Using a mammalian two-hybrid system, we have provided the first demonstration of USF1 and USF2 interaction and NRF-1 dimerization in vivo, and that USF1/USF2 heterodimers and NRF-1 interact with each other. We propose a model in which DNA bending facilitates the interaction of USF1/USF2 and NRF-1 and the juxtaposition of these factors with the basal transcription machinery. This promoter architecture may be particularly important for the regulation of genes like FMR1 which lack both a TATA-sequence and an Initiator element. Molecular basis of disease pathology in FXS "premutation" carriers: We have shown that FXS "premutation" carriers do not have elevated levels of USF1, USF2 or NRF-1. This suggests that one or more additional transcription factors may be involved in the increased FMR1 transcription in these individuals. When FMR1 reporter constructs are tested in these cells, no increased transcription is seen. This suggests that the effect on transcription might be exerted in cis, presumably due to the presence of expanded repeat tracts. This hypothesis is currently being tested. We have also shown that CGG-containing RNA readily forms very stable hairpins containing a mixture of C-G and G-G base pairs. These hairpins could have 3 important effects. Firstly, they may activate enzymes like PKR, that are sensitive to double-stranded RNA (dsRNA). This could lead to inhibition of translation and initiation of the apoptotic cascade. This may lead to the ovarian and cerebellar dysfunction seen in "premutation" carriers. Our preliminary results support this hypothesis. It has recently been suggested that dsRNA may be the trigger for epigenetic modifications of chromatin and DNA that results in transcriptional silencing. The FXS RNA hairpins may thus be responsible for the transcription initiation defect in individuals with FXS. We are currently testing this hypothesis. The hairpins may also be responsible for the stalling of the 40S ribosomal subunit that is seen on FXS transcripts, thereby producing the translation defect that is an important component of this disease. FRDA gene regulation: In individuals with the French-Acadian form of FRDA, milder symptoms are seen for any given repeat number than in individuals with the classic form of the disease. Genetics suggests that the milder symptoms are tightly linked to the FRDA locus. It may be that in these individuals additional mutations are seen that ameliorate disease progression. These mutations could include any that increase the initiation of transcription, the efficiency of splicing or stability of the transcript. We have sequenced the promoter and 3' UTR of normal individuals and individuals with the classic and French-Acadian forms of FRDA. To date we have found no mutations that affect any of these processes. We are now beginning to focus on the integrity of the repeat itself. Our previously described model for the attenuation of transcription in this disease invoked the formation of a purine:pyrimidine:purine triplex that forms during transcription and traps the RNA polymerase on the template. According to this model, interruptions to the purity of the repeat tract could reduce the stability of such a triplex, thereby leading to increased yields of full length mRNA from this gene.